by Lance Brofman
Last October we published "A Very Long-Term View Of Government Finances - Implications For The S&P 500". The forecast extended to 2088 based on macroeconomic data forecasts from the Congressional Budget Office (CBO). From these we developed alternative economic scenarios, and we forecasted profits, the S&P earnings multiple, and thus an index value for the S&P 500 (NYSEARCA:SPY).
The conclusions from that analysis were both negative and potentially positive. On the one hand CBO data forecasts revealed a very negative equity market outlook because the CBO projected that current law would result in a huge increase in government outlays relative to GDP. But on a brighter and more hopeful note we showed that the outlook could be much more positive if GDP growth were just 0.2% per year higher than the baseline CBO forecast.
In July-August this year CBO updated its budget outlook for the 2014-2024 period. We applied this updated outlook to the long term, enabling us to update near and distant budget forecasts and forecasts for profits, the earnings multiple, and the S&P index. As part of this exercise we are integrating issues raised by Dr. Lacy Hunt of Hoisington Capital Management in his recent podcast entitled "The World Economy's Terminal Debt Sclerosis".
Our accompanying Table I shows a comparison between the most recent CBO ten year budget outlook and the outlook it published last year for the same ten year period. As usual the period is forecast on the basis of current law, and the forecast assumes no recession or any meaningful externality. In its update CBO has downgraded its ten year forecast for nominal GDP from an average of 5% per year to 4.5% and for real GDP from an average 2.8% per year to 2.5%. Implicitly it has therefore downgraded its estimate of future inflation as well.
The expectation of slower economic growth weighs on government finances. While the federal budget deficit in 2014-2015 is lower than CBO's year ago forecast, the cumulative ten year deficit is now about $600 billion higher. Because GDP growth is slower and the deficit is higher, by 2024 publicly held debt as a % of GDP is at 74% versus the 70% level that was forecast in last year's report. Finally, CBO forecasts that net interest as a % of GDP is about steady relative to last year's forecast. This is owing to a lower interest rate forecast that is a by-product of slower growth and lower inflation.
This revised CBO outlook is consistent with the mosaic of the U.S. and global economy described by Lacy Hunt in his recent podcast. Lacy argues that the economy is in a vicious self- reinforcing environment of stagnation (a liquidity trap) which is an outgrowth of excess public and private sector debt. In this environment excess debt inhibits economic growth which in turn exerts downward pressure on inflation. Weak inflation in turn depresses aggregate demand, adversely affecting nominal GDP and public sector revenue growth even as the monetary authority drives down interest rates to combat slow growth and low inflation.
From Lacy's vantage point the guts of the problem is the economy's growing inability to service its debt and the fact that interest rates have not fallen sufficiently to alter the sclerotic demand environment in which the U.S. and global economy is immersed. To illustrate this conundrum Lacy utilizes a relationship codified by the Austrian School of Economics wherein the BAA interest rate is related to nominal GDP growth such that as long as the interest rate is higher than nominal GDP growth, the economy is unable to adequately service its debt, thus prolonging sluggish business activity.
The accompanying Chart I shows a coincident history between the BAA bond rate and nominal GDP growth and a five year moving average of the two for the post World War II period. Coincident data is very jagged because of periodic business expansions and contractions. The five year average smoothes this and makes the trend in the relationship more vivid. On the chart zero is the line of demarcation for debt service with a favorable level being above zero and a worsening debt service climate represented by a below zero reading. On balance whether it be coincident or smoothed data the chart shows that until 1980 the BAA rate was consistently below the GDP growth rate which is a positive for debt service. Since 1980 the opposite has prevailed even as the general level of long term interest rates declined. The debt service proxy hit a low point in 1987 and from there it improved steadily with faster economic growth in the 1990s. But even during the 1990s the economy's ability to service debt deteriorated, and of course it worsened significantly with the onset of recession in 2008-2009 and throughout the past five years of recovery.
One could identify many channels through which debt service might affect the overall economy. Its impact on business investment would be one such channel so to test this we correlated net domestic investment (the domestic capital spending of domestic companies) with the five year moving average of Lacy's debt service variable. In fact the relationship is direct, meaning the better able that debt is serviced i.e. GDP exceeds the interest rate, the more positive it is for capital spending. The accompanying Chart II shows the historical fit. The relationship is statistically significant, explaining 18% of the variation in capital spending.
The BAA bond rate seems as good a proxy as any for the economy-wide business cost of funds. It works better than the ten year treasury rate in correlations with business investment. Yet it was not as useful as the ten year treasury rate in forecasting the market multiple. We surmise this reflects the presumption that borrowing costs for 90% of S&P 500 companies are close to the ten year rate. Moreover, almost all BAA corporate bonds are callable whereas treasuries are not callable. This gives investors an asymmetry whereby if rates drop, bonds are called but if rates rise, they would most likely not be called.
With this in mind our use of the ten year treasury rate in our short and long term forecasting models of the market multiple seems quite valid. And you will recall that this variable and a forward looking moving average of federal outlays to GDP are the prime ingredients in our forecasting model of the market multiple. Our model for estimating profits includes nominal and real GDP growth and a lagged moving average of unit labor cost in manufacturing.
Last year when we estimated this model to the 2080s we found that eventually the market multiple would fall into the single digits beginning in 2034 and to zero in 2067. With newly revised input data the news is less bad as the model now shows that 2039 is when the multiple swings into single digits.
The prime factor driving down the multiple is an explosion in government's share of the economy beginning in 2025. And in the context of Lacy's analysis this explosion makes it ever more difficult to service debt. This makes a dollar's worth of earnings less valuable, leading to weaker equity prices, a rising cost of equity capital and thus ever slower investment and economic growth.
To date Central Banks have been attempting to improve the GDP-interest rate relation. Their direct method has been zero short term interest rates and quantitative easing. The goal of low interest rates and monetary expansion has been to boost aggregate demand and price inflation, implicitly in this country and explicitly in the case of Japan and more recently Europe. Lacy would argue that while these Central Bank policies are effective intermittently, the effect is transitory and they are doomed to fail.
We are sympathetic to this argument as we have long argued that the U.S. and the global economy is caught in a liquidity trap. Of course Central Bank policies would be more effective if fiscal policies were in sync. But the fact is that over regulation, and tax and spending policies have often been countervailing. Given this we have also long argued that what is needed to break the debt spiral is some new innovation that would spur demand independently of public policies. Growth was facilitated in the 1950s and 1960s with the construction of the interstate highway system and the space race. Thus, the economy was very able to service the debt that was incurred during World War II.
Growth was facilitated in the 1990s by the advent of the internet and advances in computer technology. The 1990s was a rapid growth era even though Lacy's debt service ratio was consistently below zero. This implies that while debt service may be an impediment to growth it certainly is not the only factor affecting economic growth. The same is implied by the correlation between debt service and capital spending. The relation is significant but not conclusive.
The problem is that as debt service is negative and it is allowed to fester, the impact eventually becomes debilitating. This becomes clear when examining the very long term outlook to the 2080s. As noted earlier, upon running our models for the very long term we found that federal outlays to GDP rose to such a degree that it drove the earnings multiple to zero in the 2060s. With newly revised data the news is slightly less bad in that it is not until 2074 that the multiple falls to zero. This we would describe as providing cold comfort at best.
The accompanying Table II shows this and accompanying CBO macro forecast data to 2089. Various measures remain at tolerable levels until the middle of the next decade when they begin to explode. For example, the federal deficit would reach 12.7% of GDP at the terminal date versus about 3% currently and 14.7% in CBO's earlier analysis. Debt held by the public would reach 229% of GDP by 2089 versus 74% currently and 245% in the analysis done last year.
The major drivers of future deficits have not changed either in size or in order of importance. Federally financed health care and interest payments on the federal debt are the two most significant with social security outlays coming in a distant third. Federal health care spending is shown to rise from roughly 4.9% of GDP this year to 14% at the end of the forecast period in 2089. Of the total, spending on Medicare is shown to rise from 3% to 9% of GDP while spending on medical, child health insurance, and Affordable Care Act subsidies rise from 1.9% to 4.7% of GDP.
Net interest payments rise even as interest rates are lower than initially forecast by CBO. Interest payments are projected to rise from 1.3% of GDP currently to 10% of GDP by the end of the forecast period. Of course this increase parallels the rise in the amount of debt held by the public. Net interest payments are untouchable-the obligations have to be met-and there certainly does not and probably will not be any political will to halt the rise in federal health care spending as the population continues to age.
Ominously one cannot convincingly argue that this is even a worst case. Indeed, CBO projects that all federal spending excluding health care and net interest will actually decline from 14.2% of GDP currently to 12.6% of GDP eventually. This is actually a sharper decline than CBO forecast for this category in last year's document. Moreover, considering that social security is in this catchall category and is projected to rise from 4.7% of GDP to 6.9%, the reductions elsewhere are even sharper. Defense is in this category and it is hard to convincingly argue that the world has become a safer place.
Finally, regarding tax policy CBO projects a revenue rise from 17.6% of GDP currently and in the vicinity of the long term average to 23.9% by 2089 or considerably above the long term average. Raising the ratio above the 18.5% average seems very difficult to accomplish in the short term and very likely the long term as well. If Dr. Lacy Hunt is currently fearful of the consequences of the economy's current debt service predicament, he has to be absolutely terrified at the long run outlook that is forecast y CBO under current law.
But a financial disaster is not inevitable. We demonstrated in last year's Very Long Term Outlook how significant an economic growth path that was merely 0.2% per year higher than the CBO forecast could be. Indeed the federal budget deficit would disappear by the mid-2080s and the debt to GDP ratio would shrink to a very manageable 45% in a permanently slightly faster growing economy. With CBO newly revised estimates budget balance could be achieved ten years earlier. Thus, the secular stagnation thesis of Dr. Larry Summers would prove transitory as opposed to secular and Lacy would sleep better at night as the economy's ability to service its debt improved.
Achieving faster economic growth through some combination of faster real activity and higher inflation may be easier said than done. The optimal vehicle for achieving a faster GDP growth path will continue to be a matter of acrimonious debate. But while action seems elusive there does seem to be an emerging consensus that the tax and regulatory apparatus needs to be overhauled and spending needs to be more efficient. This will not be easy or quick and in the meanwhile incentives for growth oriented innovation need to be emphasized in order to boost aggregate demand and propel the economy out of its liquidity trap.
(click to enlarge)
Chart II Actual and Predicted - Net domestic investment: Private: Domestic business as a ratio of GDP 5-year moving averages 1964-2013 as a function of the difference between GDP growth and Baa corporate bond yields
(click to enlarge)
Source: Growth Debt And Secular Stagnation - Implications For The S&P 500
Additional disclosure: Please note that this article was written by Dr. Vincent J. Malanga and Dr. Lance Brofman with sponsorship by BEACH INVESTMENT COUNSEL, INC. and is used with the permission of both.
See the original article >>